Are these FC or regular disks ? And when you say they disappear what exactly does that mean ? Does SVM complain that could start because mddbs failed ? Are the entries there in /kernel/drv/md.conf ?
-Sanjay Steve Peng wrote: > Lars Tunkrans wrote: > >> Hi, >> I experienced a disasterous lvm installation failure today. On >> solaris 9 9/05 SPARC V240 with 2 * 73 Gb disk and 2 * 146 GB disk >> By placing the metadb in the last 128 MB on the " new" 146 GB >> disks on a SUNFIRE V240, and re-booting the server . When LVM >> tries to activate itself on powerup , all metadb copies vanish >> into the great bit-bucket never to be seen again. >> I thought about this for awhile ... Then I reinstalled the >> metadbs again on only my two 73 GB disk. this time lvm behaved >> normally. >> Question : is LVM vunerable if metadb's are placed after the >> 137 GB boundary on a disk I will try next week to place the >> metadbs first on the 146 GB disk and see if they are stable in that >> configuration. >> Regards >> //Lars >> This message posted from opensolaris.org >> _______________________________________________ >> lvm-discuss mailing list >> lvm-discuss at opensolaris.org >> >> > It will be interesting to know what you find out. It really doesn't > matter where > you put the mddb as long as SVM can access the partition. Did you see > any disk error during the reboot? You can check out /var/adm/messages > or run dmesg. > > Steve > _______________________________________________ > lvm-discuss mailing list > lvm-discuss at opensolaris.org
